Tracegate CI runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 1 Nov 2026.
Local CI broker and test-impact ledger for coding agents: queue, lane modes and red-run packets.
Today is the operative word: we check Tracegate CI every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add tracegate-ci -- npx -y tracegate-ci
{
"mcpServers": {
"tracegate-ci": {
"command": "npx",
"args": [
"-y",
"tracegate-ci"
]
}
}
}
[mcp_servers.tracegate-ci]
command = "npx"
args = ["-y", "tracegate-ci"]
{
"mcpServers": {
"tracegate-ci": {
"command": "npx",
"args": [
"-y",
"tracegate-ci"
]
}
}
}
{
"servers": {
"tracegate-ci": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"tracegate-ci"
]
}
}
}
Errors, traces, replays and logs from Talaria for coding agents.
Time tracking for agencies: log time, run the timer, check client budgets and pull reports.
Read-only MCP server for coding-agent traces, alerts, costs, usage, and budget health.
Accessibility-tree browser for AI agents: read pages, log in, extract data, run tasks.
Read local agent-run JSONL traces through MCP for failures, approvals, and run comparisons.
Audit trail and traceability for AI agents. MCP gate: evaluate, ingest, request approval.
Sendora's docs and API for coding agents: send test mail, read the log, set up webhooks.
Open eval leaderboard + CI gate for autonomous coding agents (solve, score, trace).
Answers built from our own checks of this server.